Google’s Gemini 3.8 Flash in Make: handle text, images, audio, and video in automation workflows
When a scenario needs to extract details from a document, image, audio recording, or video, Gemini 3.8 Flash can accept those inputs and return text for the next workflow step.
Gemini 3.8 Flash is Google’s model. You can automate your workflows with it in Make using Google Gemini AI app, Make AI Toolkit app, and Make AI Agents.
What’s new in Gemini 3.8 Flash
Improved workhorse model: Google reports improvements over Gemini 3.7 Flash in software engineering, agentic tasks, and critical multi-step reasoning in specialized domains.
Configurable thinking: Choose LOW, MEDIUM, or HIGH thinking levels for a task. MEDIUM is the default.
Structured and tool-based tasks: The model supports structured output, function calling, and code execution, subject to each capability’s launch stage.
Large context capacity: It supports a 1,048,576-token context window and up to 65,536 output tokens.
Gemini 3.8 Flash use cases for Make automation
Automate invoice and contract intake: Send a scanned PDF or photographed invoice to Gemini 3.8 Flash, extract line items and totals as structured output, then route the parsed data straight into your accounting or ERP module.
Triage support tickets from screen recordings: Have Gemini 3.8 Flash watch a customer’s screen recording, summarize the issue in text, and auto-tag and route the ticket to the right team based on severity.
Log sales calls to your CRM: Feed a call recording to Gemini 3.8 Flash, get a structured summary of key points, action items, and sentiment, and append it directly to the matching CRM record.
Run a multi-tool research agent: Use Make AI Agents with function calling and code execution to pull data from multiple connected tools, run a calculation, and produce a final structured report.- Cache recurring reference material: Use context caching so Gemini 3.8 Flash doesn’t reprocess the same reference document on every run, cutting latency and token cost in high-frequency scenarios.
Ground generated content before publishing: Use URL context and grounding to verify AI-generated copy or support answers against a live webpage, flagging factual mismatches before they go out.
Gemini 3.8 Flash requires a Gemini API key in Make, and higher thinking effort can use more tokens.
How to use Gemini 3.8 Flash in Make
Google’s Gemini 3.8 Flash is available in these Make native apps with your Gemini API key:
Make AI Toolkit for quick setup.
Google Gemini AI for structured prompts with your provider connection.
Make AI Agents for agentic workflow automations.
Try my starter scenario to see the latest Gemini AI models automated with Make: https://ma.ke/gemini-ai-demo
Gemini 3.8 Flash helps you move information from text, images, audio, and video into workflow steps, while choosing the thinking level that fits each task.
Happy automating! 
Valery from Make
Helpful resources